HELP.. I need to do this fast :$

def get_winner (dict_winner):

new_dict = {}
for winner in dict_winner:
first_letter = winner[0]
value = dict_winner[winner]
if first_letter in new_dict:
new_dict[first_letter] += value
else:
new_dict[first_letter] = value
return (new_dict)



>>> get_winner({('C', 'A', 'B', 'D') :3, ('D', 'B', 'C', 'A') :2, ('C', 'D', 'A', 'B') :1, ('A', 'D', 'B', 'C'):2, ('A', 'D', 'C', 'B') :4, ('A', 'C', 'D', 'B') :2})

Output =
{'A': 8, 'D': 2, 'C': 4}

But i want it to give me a final output of a tuple of str, NoneType..

Also.. it is eliminating only the letter with the smallest value in first place only one time.. i want it to repeat this process until i get one winner in the end.. so in this case.. All the B's will be eliminated in the dict itself, not in the out put.
for example:

first time = [8, 0, 4, 2]
second time =
{('C', 'A', 'D') :3,
('D', 'C', 'A') :2,
('C', 'D', 'A') :1,
('A', 'D', 'C'):2,
('A', 'D', 'C') :4,
('A', 'C', 'D') :2})
C = 4
D = 2
A = 8
third time=
{('C', 'A') :3,
('C', 'A') :2,
('C','A') :1,
('A', 'C'):2,
('A', 'C') :4,
('A', 'C') :2})
C = 6
A = 8

8/ 14 = more than 50%, I know that should have been the case since the beginning because A already had the majority value. but i am assuming A has a value of 40% which is when elimination should begin.. So can you point out where i went wrong in coding this?... which makes A the winner!!
And my output should be --

('A', None)

I tried your program, getting nowhere fast.

I think there was an indentation problem:
this should work: